This project investigates the role of cerebral imaging in prognosis of stroke recovery by evaluating how blood flow and NMR spectroscopic imaging methods can separate and quantify structural, vascular, and neuro-metabolic abnormalities post-stroke. This has potential significance to provide a rational, base on physiological measurements, for patient management and rehabilitation strategies, including the use of facilities versus compensatory rehabilitation techniques. The primary emphasis is on identifying the diaschisis component of reduced activity in the stroke penumbra, defined as a reduction in regional neurometabolic activity outside the infarct not due to hemodynamic vascular constraints, and presumably due to differentiation effects of the stroke. Since the amount of non-infarcted neural tissue identifies the amounts of affected brain that has potential for recovery, it is hypothesized that volumetric quantitation of diaschisis will be highly correlated with stroke recovery prognosis. The diaschisis component will be identified by determining the volume of brain tissue with reduced blood flow that is in excess of the measured anatomic infarction size. The excessive volume (which is not due to ischemia) is presumed to be due to the reactive reduction of metabolism from diaschisis. Neuro-metabolic changes caused by combined diaschisis and ischemia will also be studies. T1 and T2 MRI will be conducted to determine the volume of brain outside the infarction to anatomically localize the penumbra. Brain SPECT and 4.1T NMR spectroscopic imaging (MRSI) will be performed to determine the vascular and metabolic status of the extended stroke penumbra. These measurements will be conducted in cerebrovascular disease patients and 30 controls to provide independent measures of r-CBF, r-CBF reserve and reactivity, and regional metabolic activity as reflected by 1H and 31P NMR spectroscopy. Ischemia or any vascular constraints on recovery will be determined via Tc-99m HM-PAO SPECT measures of resting r-CBF and r-CBF reactivity to acetazolamide stress . Corroborative, biochemical evidence for the nature of any focal reductions in r-CBF activity will come from MRSI measures of lactate, choline, creatine, N-acetyl aspartate, glutamate, inorganic phosphate, ATP, and phosphocreatine. Together the measure will provide a way to identify and quantitate recoverable components of stroke efforts. All functional, anatomic and metabolic neuroimaging procedures will be conducted at two-week, six-week, six-month, and one year time intervals post-stroke. Neurological, behavioral, and cognitive status evaluations will be conducted at the same intervals. Outcome at one year post stroke will be compared with the temporally obtained imaging parameters to evaluate which parameters provide the most accurate prognosis of recovery. A preliminary evaluation will also be made of the extent to which recovery of function may involve relearning by brain regions not affected by the stroke, as compared the resolution of diaschisis in the affected region.